COVID-19 Teşhisinde Makine Öğrenmesi Modellerinin Performansını Artırmaya Yönelik SHAP Analizi ile Desteklenen Hiperparametre Optimizasyonu

Abstract

SARS-CoV-2’nin ortaya çıkışı, etkili tanı araçlarının geliştirilmesine yönelik bilimsel çalışmalarda artışa yol açmıştır. Salgının kontrol altına alınabilmesi için doğru teşhis büyük önem taşımakta olup, yapay zeka (YZ) tabanlı yöntemler bu alanda umut vadetmektedir. Bu çalışmada, COVID-19’un kan değerlerinden, özellikle de Van Yüzüncü Yıl Üniversitesi Dursun Odabaş Tıp Merkezi’nden elde edilen hemogram test sonuçlarından, makine öğrenmesi (ML) teknikleri ile tahmin edilmesi amaçlanmıştır. Çeşitli ML algoritmaları test edilmiş ve en yüksek doğruluk oranı Rastgele Orman (Random Forest) yöntemiyle elde edilmiştir. Modelin performansı, optimizasyon süreciyle daha da artırılmış; bu süreçte Genetik Algoritma (GA) en etkili yöntem olarak öne çıkmıştır. Modelin kararlarını etkileyen temel özellikleri belirleyerek yorumlanabilirliği artırmak amacıyla SHAP analizi uygulanmıştır. Değerlendirilen üç veri seti arasında, en yüksek doğruluk oranı (%91,56) Veri Seti 3’te elde edilmiştir. Optimizasyon sonrası Veri Seti 2 dengeli bir performansla %85,09 doğruluk oranına ulaşırken, Veri Seti 1’de doğruluk %65,02’ye yükselmiş ancak duyarlılık (recall) düşüktür. GA ile optimize edilen model, 0.9467 AUC değerine ulaşarak güçlü bir sınıflandırma başarısı göstermiştir. Bu bulgular, hastalık tespitinde YZ destekli modellerin etkinliğini ve sağlık sistemlerini daha hızlı ve doğru teşhis imkânı sunarak destekleme potansiyelini ortaya koymaktadır. Gelecek çalışmalarda, farklı modelleme stratejileri ve derin öğrenme tekniklerinin entegrasyonuyla tanı doğruluğunun daha da artırılması hedeflenmektedir.

Keywords

• Hemogram
• COVID-19
• Rastgele orman
• Genetik algoritma
• SHAP analizi

Hyperparameter Optimization Supported by SHAP Analysis for Performance Enhancement of Machine Learning Models in Covid-19 Diagnosis

Abstract

The emergence of SARS-CoV-2 has led to increased scientific focus on developing effective diagnostic tools. Accurate detection is crucial for controlling the outbreak, and artificial intelligence (AI)-based methods have shown promise. This study uses machine learning (ML) techniques to predict COVID-19 from blood values, specifically, hemogram test results obtained from Van Yuzuncu Yil University Dursun Odabas Medical Center. Various ML algorithms were tested, with the Random Forest method achieving the highest accuracy. Model performance was further improved through optimization, where the Genetic Algorithm (GA) proved most effective. SHAP analysis was employed to enhance the interpretability of the predictions by identifying key features influencing the model’s decisions. Among the three evaluated datasets, Dataset 3 achieved the highest accuracy (91.56%). Dataset 2, after optimization, reached 85.09% accuracy with balanced performance, while Dataset 1 saw improved accuracy (65.02%) but lower recall. The GA-optimized model reached an AUC of 0.9467, indicating strong classification capability. These findings highlight the effectiveness of AI-driven models in disease detection and their potential to support healthcare systems by enabling faster and more accurate diagnosis. Future efforts will focus on integrating different modeling strategies and deep learning techniques to further improve diagnostic accuracy.

Keywords

• Hemogram
• COVID-19
• Random forest
• Genetic algorithm
• SHAP analysis

References

1. Raji P., & Lakshmi, G. R. D. (2020). COVID-19 pandemic analysis using regression. medRxiv. https://doi.org/10.1101/2020.10.08.20208991
2. Bandil, S., & Rathore, S. (2019). Study of haematological parameters in malaria. Asian Journal of Medical Research, 8(3), PT08-PT12. https://doi.org/10.21276/ajmr.2019.8.3.pt3
3. Demircan, S. (2022). Öksürük sesi kayıtlarından spektral özellikler ile otomatik COVID-19 tespiti. Avrupa Bilim ve Teknoloji Dergisi, Ejosat Special Issue 2022 (ICAENS-1), 492-495. https://doi.org/10.31590/ejosat.1083052
4. Demir, F. B., & Yılmaz, E. (2021). X-ray görüntülerinden COVID-19 tespiti için derin öğrenme temelli bir yaklaşım. Avrupa Bilim ve Teknoloji Dergisi, Ejosat Special Issue 2021 (RDCONF), 627-632. https://doi.org/10.31590/ejosat.1039522
5. Taş, F., Özüdoğru, O., & Bolatlı, G. (2020). Bilgisayarlı tomografi bulguları negatif olan COVID-19 hastalarının; epidemiyolojik, klinik ve laboratuvar sonuçları açısından değerlendirilmesi. Selçuk Sağlık Dergisi, COVID-19 Özel, 18-32. Retrieved from https://dergipark.org.tr/en/pub/ssd/issue/57170/757740
6. Ayata F, Seyyarer E. (2024). COVID-19 Diagnosis from Blood Gas Using Multivariate Linear Regression. Hittite J Sci Eng. 2024;11(1):15-23.
7. Ayata, F. (2024). Covıd-19 Tespiti ve Salgın Yönetiminde Makine Öğrenmesi: Kan Gazı Analizine Dayalı Bir Yaklaşım. Doğu Fen Bilimleri Dergisi, 7(1), 1-10. https://doi.org/10.57244/dfbd.1492816
8. Dülger, D., & Ekici, S. (2020). Günümüz pandemisi COVID-19’un laboratuvar tanı yöntemleri. Avrasya Sağlık Bilimleri Dergisi, COVID-19 Special Issue, 111-115. Retrieved from https://dergipark.org.tr/en/pub/avrasyasbd/issue/56010/755340

9. Tezcan Keleş, G., & Bozkurt, İ. (2021). COVID-19 Hastalığı Tanı ve Tedavisinde Kullanılan Laboratuvar Testleri. Celal Bayar Üniversitesi Sağlık Bilimleri Enstitüsü Dergisi, 8(2), 380-387. https://doi.org/10.34087/cbusbed.865763
10. Dokeroglu, T., Sevinc, E., Kucukyilmaz, T., & Cosar, A. (2019). A survey on new generation metaheuristic algorithms. Computers & Industrial Engineering. https://doi.org/10.1016/j.cie.2019.106040
11. Villarrubia, G., De Paz, J. F., Chamoso, P., & De la Prieta, F. (2018). Artificial neural networks used in optimization problems. Neurocomputing. https://doi.org/10.1016/j.neucom.2017.04.075
12. Pandey, M., Zakwan, M., Sharma, P. K., & Ahmad, Z. (2020). Multiple linear regression and genetic algorithm approaches to predict temporal scour depth near circular pier in non-cohesive sediment. ISH Journal of Hydraulic Engineering. https://doi.org/10.1080/09715010.2018.1457455
13. Shihabudheen, K. V., Mahesh, M., & Pillai, G. N. (2018). Particle swarm optimization based extreme learning neuro-fuzzy system for regression and classification. Expert Systems with Applications. https://doi.org/10.1016/j.eswa.2017.09.037
14. Liu, Z., Fan, G., Sun, D., Wu, D., Guo, J., Zhang, S., Yang, X., Lin, X., & Ai, L. (2022). A novel distributed energy system combining hybrid energy storage and a multi-objective optimization method for nearly zero-energy communities and buildings. Energy, 239(Part E), 122577. https://doi.org/10.1016/j.energy.2021.122577
15. Oyelade, O. N., Ezugwu, A. E.-S., Mohamed, T. I. A., & Abualigah, L. (2022). Ebola optimization search algorithm: A nature-inspired novel metaheuristic optimization algorithm. IEEE Access, 10, 16150-16177. https://doi.org/10.1109/ACCESS.2022.3147821
16. AlJame, M., Ahmad, I., Imtiaz, A., & Mohammed, A. (2020). Ensemble learning model for diagnosing COVID-19 from routine blood tests. Informatics in Medicine Unlocked, 21, 100449. https://doi.org/10.1016/j.imu.2020.100449.
17. Alves, M. A., Zanon de Castro, G., Soares Oliveira, B. A., Ferreira, L. A., Ramírez, J. A., Silva, R., & Guimarães, F. G. (2021). Explaining Machine Learning based Diagnosis of COVID-19 from Routine Blood Tests with Decision Trees and Criteria Graphs. Computers in Biology and Medicine, 132, 104335. https://doi.org/10.1016/j.compbiomed.2021.104335.
18. Meng, Z., Wang, M., Song, H., Guo, S., Zhou, Y., Li, W., Zhou, Y., Li, M., Song, X., Zhou, Y., Li, Q., Lu, X., & Ying, B. (2020). Development and utilization of an intelligent application for aiding COVID-19 diagnosis. Içinde medRxiv (s. 2020.03.18.20035816). medRxiv. https://doi.org/10.1101/2020.03.18.20035816.
19. Batista, A. F. de M., Miraglia, J. L., Donato, T. H. R., & Chiavegatto Filho, A. D. P. (2020). COVID-19 diagnosis prediction in emergency care patients: A machine learning approach. medRxiv (s. 2020.04.04.20052092). medRxiv. https://doi.org/10.1101/2020.04.04.20052092.
20. Avila E, Kahmann A, Alho C, Dorn M. Hemogram data as a tool for decision-making in COVID-19 management: applications to resource scarcity scenarios. PeerJ. 2020 Jun 29; 8: e9482. doi: 10.7717/peerj.9482. PMID: 32656001; PMCID: PMC7331623.
21. Feltes BC, Vieira IA, Parraga-Alava J, Meza J, Portmann E, Terán L, Dorn M. Feature selection reveal peripheral blood parameter's changes between COVID-19 infections patients from Brazil and Ecuador. Infect Genet Evol. 2022 Mar; 98: 105228. doi: 10.1016/j.meegid.2022.105228. Epub 2022 Jan 30. PMID: 35104680; PMCID: PMC8800568.
22. Bao, F. S., He, Y., Liu, J., Chen, Y., Li, Q., Zhang, C. R., Han, L., Zhu, B., Ge, Y., Chen, S., Xu, M., & Ouyang, L. (2020). Triaging moderate COVID-19 and other viral pneumonias from routine blood tests. arXiv. http://arxiv.org/abs/2005.06546.
23. Yang, H. S., Hou, Y., Vasovic, L. V., Steel, P., Chadburn, A., Racine-Brzostek, S. E., Velu, P., Cushing, M. M., Loda, M., Kaushal, R., Zhao, Z., & Wang, F. (2020). Routine laboratory blood tests predict SARS-CoV-2 infection using machine learning. medRxiv (s. 2020.06.17.20133892). medRxiv. https://doi.org/10.1101/2020.06.17.20133892
24. Sun, N. N., Yang, Y., Tang, L. L., Dai, Y. N., Gao, H. N., Pan, H. Y., & Ju, B. (2020). A prediction model based on machine learning for diagnosing the early COVID-19 patients. medRxiv (s. 2020.06.03.20120881). medRxiv. https://doi.org/10.1101/2020.06.03.20120881
25. Seyyarer, E., Karci, A. & Ateş, A. (2022). Stokastik ve deterministik hareketlerin optimizasyon süreçlerindeki etkileri. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 37 (2), 949-966. DOI: 10.17341/gazimmfd.887976
26. Elaziz, M. A., Hosny, K. M., Salah, A., Darwish, M. M., Lu, S., & Sahlol, A. T. (2020). New machine learning method for image-based diagnosis of COVID-19. PLoS One, 15(6), e0235187. https://doi.org/10.1371/journal.pone.0235187
27. Agushaka, J.O.; Ezugwu, A.E.; Abualigah, L. (2022). Dwarf Mongoose Optimization Algorithm. Comput. Methods Appl. Mech. Eng. 391, 114570
28. Ramachandran, M., Mirjalili, S., Nazari-Heris, M., Parvathysankar, D. S., Sundaram, A., & Charles Gnanakkan, C. A. R. (2022). A hybrid Grasshopper Optimization Algorithm and Harris Hawks Optimizer for Combined Heat and Power Economic Dispatch problem. Engineering Applications of Artificial Intelligence, 111, [104753]. https://doi.org/10.1016/j.engappai.2022.104753
29. Katoch S, Chauhan SS ve Kumar V. A review on genetic algorithm: past, present, and future. vol. 80. Multimedia Tools and Applications; 2021. https://doi.org/10.1007/s11042-020-10139-6.
30. Sevinçli M. (2023). İki Robot Kollu Sistemde Görüntü İşleme Ve Genetik Algoritma Kullanarak Çarpışma Önleme ve Görev Paylaşımının Optimizasyonu. Yüksek Lisans Tezi, Elektrik-Elektronik Mühendisliği Anabilim Dalı. İskenderun Teknik Üniversitesi.
31. Çavuşlu, M. A., Karakuzu, C. & Şahin, S. (2010). Parçacık Sürü Optimizasyonu Algoritması ile Yapay Sinir Ağı Eğitiminin FPGA Üzerinde Donanımsal Gerçeklenmesi. Politeknik Dergisi, 13 (2), 83-92. Retrieved from https://dergipark.org.tr/en/pub/politeknik/issue/33053/367865
32. Bergstra, J., & Bengio, Y. (2012). Random Search for Hyper-Parameter Optimization. Journal of Machine Learning Research, 13, 281-305.
33. Hutter, F., Kotthoff, L., & Vanschoren, J. (2019). Automated Machine Learning: Methods, Systems, Challenges. Springer.
34. Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., & Talwalkar, A. (2017). Hyperband: A novel bandit-based approach to hyperparameter optimization. Journal of Machine Learning Research, 18, 1-52.
35. Snoek, J., Larochelle, H., & Adams, R. P. (2012). Practical Bayesian Optimization of Machine Learning Algorithms. Advances in Neural Information Processing Systems, 25, 2951–2959.
36. Boser, B. E., Laboratories, T. B., Guyon, I. M., Laboratories, T. B., & Vapnik, V. N. (1992). SVM-A training algorithm for optimal margin classifiers.pdf.
37. Khorraminezhad, L., Leclercq, M., Droit, A., Bilodeau, J. F., & Rudkowska, I. (2020). Statistical and machine-learning analyses in nutritional genomics studies. Içinde Nutrients (C. 12, Sayı 10, ss. 1–19). MDPI AG. https://doi.org/10.3390/nu12103140.
38. Farnaaz, N., & Jabbar, M. A. (2016). Random Forest Modeling for Network Intrusion Detection System. Procedia Computer Science, 89, 213–217. https://doi.org/10.1016/j.procs.2016.06.047.
39. Korkem, E. (2013). Mikroarray Gen Ekspresyon Veri Setlerinde Random Forest Ve Naıve Bayes Sınıflama Yöntemleri Yaklaşımı. Hacettepe Üniversitesi.
40. Martínez Torres, J., Iglesias Comesaña, C., & García-Nieto, P. J. (2019). Review: machine learning techniques applied to cybersecurity. International Journal of Machine Learning and Cybernetics, 10(10), 2823–2836. https://doi.org/10.1007/s13042-018-00906-1.
41. Abraham, A. (2005). Artificial Neural Networks. Handbook of Measuring System Design, 901-908.
42. Köse, U. ve Arslan, A. (2017). Optimization of Self-Learning in Computer Science Engineering Course: An İntelligent Software System Supported by Artifcial Neural Network and Vortez Optimization Algorithm. Comput Appl Eng Educ 25, 142–156.
43. Kılınç, D., Borandağ, E., Yücalar, F., Tunalı, V., Şimşek, M., & Özçift, A. (2016). KNN Algoritması ve R Dili ile Metin Madenciliği Kullanılarak Bilimsel Makale Tasnifi. Marmara Fen Bilimleri Dergisi, 28(3), 89-94.
44. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
45. Güler, E. (2023). XGBoost Algoritması. Medium. Retrieved from https://efecanxrd.medium.com/xgboost-algoritmas%C4%B1-6703b14efd5 (Accessed: February 4, 2025).
46. Ultralytics. (2023). XGBoost. Ultralytics Documentation. Retrieved from https://www.ultralytics.com/tr/glossary/xgboost (Accessed: February 4, 2025).
47. Sadr, H., Salari, A., Ashoobi, M. T., & Nazari, M. (2024). Cardiovascular disease diagnosis: a holistic approach using the integration of machine learning and deep learning models. European Journal of Medical Research, 29(1), 455.
48. Kaggle (2020), Einstein Data4u. Accessed 22 November 2024, https://www. kaggle. com/ einst einda ta4u/ covid 19/ versi on/4
49. Ordoñez-Avila R, Parraga-Alava J, Hormaza JM, Vaca-Cárdenas L, Portmann E, Terán L, Dorn M. CBCovid19EC: A dataset complete blood count and PCR test for COVID-19 detection in Ecuadorian population. Data Brief. 2023 Mar 1;47:109016. doi: 10.1016/j.dib.2023.109016. PMID: 36942101; PMCID: PMC10023941.
50. Guan, W. J., Ni, Z. Y., Hu, Y., Liang, W. H., Ou, C. Q., He, J. X., ... & Zhong, N. S. (2020). Clinical characteristics of coronavirus disease 2019 in China. New England journal of medicine, 382(18), 1708-1720.
51. Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., Hu, Y., ... & Cao, B. (2020). Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The lancet, 395(10223), 497-506.
52. Barnes, B. J., Adrover, J. M., Baxter-Stoltzfus, A., Borczuk, A., Cools-Lartigue, J., Crawford, J. M., ... & Egeblad, M. (2020). Targeting potential drivers of COVID-19: Neutrophil extracellular traps. Journal of Experimental Medicine, 217(6), e20200652.
53. Qin, C., Zhou, L., Hu, Z., Zhang, S., Yang, S., Tao, Y., ... & Tian, D. S. (2020). Dysregulation of immune response in patients with coronavirus 2019 (COVID-19) in Wuhan, China. Clinical infectious diseases, 71(15), 762-768.
54. Lagunas‐Rangel, F. A. (2020). Neutrophil‐to‐lymphocyte ratio and lymphocyte‐to‐C‐reactive protein ratio in patients with severe coronavirus disease 2019 (COVID‐19): a meta‐analysis. Journal of medical virology, 92(10), 1733.
55. Müller, L., Andrée, M., Moskorz, W., Drexler, I., Walotka, L., Grothmann, R., ... & Schaal, H. (2021). Age-dependent immune response to the Biontech/Pfizer BNT162b2 coronavirus disease 2019 vaccination. Clinical Infectious Diseases, 73(11), 2065-2072.
56. Tanni, F., Akker, E., Zaman, M. M., Figueroa, N., Tharian, B., & Hupart, K. H. (2020). Eosinopenia and COVID-19. Journal of Osteopathic Medicine, 120(8), 504-508.
57. Bayramoğlu, B., & Tayfur, İ. (2024). The Importance of Eosinophils in the Course of Diseases: Eosinopenia, COVID-19 and Mortality. HAYDARPAŞA NUMUNE MEDICAL JOURNAL, 64(4), 466-471.